Scale Issues in Soil Hydrology Related to Measurement and Simulation: A Case Study in Colorado.

State variables, such as soil water content (SWC), are typically measured or inferred at very small scales while being simulated at larger scales relevant to spatial management or hillslope areas. Thus there is an implicit spatial disparity that is often ignored.  Surface runoff, on the other hand, is typically measured only at a watershed outlet or edge of a field, while distributed models may simulate runoff at discrete locations in space and time. Even if water dynamics are only estimated at the whole watershed level, spatial variability of soil properties can affect responses at larger scales.  Here, we integrate previous scaling work based on measurements in an agricultural field in eastern Colorado with new measurements and simulations across scales within a 56-ha watershed defined by a drop-box (self-scouring) flume at the field boundary.  New measurements include SWC and snow water equivalent inferred at a scale of about 20 ha using a Cosmic Ray Probe (CRP).  Preliminary evaluation of CRP data includes comparisons with spatial point samples of snow depth and water equivalent after two snow events, and later samples of SWC in soil cores to 60 cm depth together with TDR measurements at 0-15 cm and 0-30 cm depth intervals.  SWC is also inferred hourly using capacitance sensors at multiple landscape positions and depth intervals of 25-35 cm and 55-65 cm.  Soil hydrology is simulated under dryland wheat-fallow rotations in alternating strips, then under a conservation program with perennial vegetation (grasses, alfalfa, and various remnant weeds during the conversion).  The AgroEcoSystem (Ages) model was developed and used to simulate distributed soil-plant-water interactions.  Ages is applied at different spatial resolutions to simulate some of the scale effects from smaller to larger land areas.  Results are compared with spatially averaged point measurements of SWC and CRP-based estimates at simulated scales.